An Iterative Equalization and Decoding Approach for Underwater Acoustic Communication

In this paper, we present an iterative approach for recovering information sent over a shallow underwater acoustic (UWA) communication channel. The procedure has three main tasks: estimation of channel model parameters (CMPs), channel equalization, and decoding. These tasks are performed cyclicly until the algorithm converges. Information bits are convolutionally encoded, punctured and permuted, mapped into quaternary phase-shift keying (QPSK) symbols, linearly modulated, and transmitted through a downward-refracting ocean waveguide. Training symbols are prepended to the transmitted sequence for initial estimation of CMPs. Our algorithm processes data from a single receive sensor. Data are received on a vertical array and the performance of the algorithm for each sensor in the array is examined. There is negligible Doppler spread in the received data. However, difference between transmitter and receiver clocks as well as slight motion of the receive array produce a nonnegligible compression of the received signals. Consequently, there is observable Doppler ldquoshift.rdquo Nonuniform resampling of the data produces time series we model as the output of a linear time-invariant system. Resampling and CMP estimation are done iteratively, in conjunction with equalization and decoding. The algorithm successfully processes the data to yield few or no information bit errors.